Anionic Effect on Electrical Transport Properties of Solid Co2+/3+ Redox Mediators

In a solid-state dye-sensitized solar cell, a fast-ion conducting (σ25°C > 10−4 S cm−1) solid redox mediator (SRM; electrolyte) helps in fast dye regeneration and back-electron transfer inhibition. In this work, we synthesized solid Co2+/3+ redox mediators using a [(1 − x)succinonitrile: x poly(ethylene oxide)] matrix, LiX, Co(tris-2,2′-bipyridine)3(bis(trifluoromethyl) sulfonylimide)2, and Co(tris-2,2′-bipyridine)3(bis(trifluoromethyl) sulfonylimide)3 via the solution-cast method, and the results were compared with those of their acetonitrile-based liquid counterparts. The notation x is a weight fraction (=0, 0.5, and 1), and X represents an anion. The anion was either bis(trifluoromethyl) sulfonylimide [TFSI−; ionic size, 0.79 nm] or trifluoromethanesulfonate [Triflate−; ionic size, 0.44 nm]. The delocalized electrons and a low value of lattice energy for the anions made the lithium salts highly dissociable in the matrix. The electrolytes exhibited σ25°C ≈ 2.1 × 10−3 (1.5 × 10−3), 7.2 × 10−4 (3.1 × 10−4), and 9.7 × 10−7 (6.3 × 10−7) S cm−1 for x = 0, 0.5, and 1, respectively, with X = TFSI− (Triflate−) ions. The log σ–T−1 plot portrayed a linear curve for x = 0 and 1, and a downward curve for x = 0.5. The electrical transport study showed σ(TFSI−) > σ(Triflate−), with lower activation energy for TFSI− ions. The anionic effect increased from x = 0 to 1. This effect was explained using conventional techniques, such as Fourier transform infrared spectroscopy (FT-IR), X-ray diffractometry (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), UV–visible spectroscopy (UV-vis), differential scanning calorimetry (DSC), and thermogravimetric analysis (TGA).


Introduction
Solar cells convert sunlight energy into electrical energy without producing harmful greenhouse gases, which is one of the crucial factors in climate change [1,2].Amongst various types of solar cells, only solid-state dye-sensitized solar cells (DSSCs) are beneficial in high-temperature regions such as Gulf countries [3].The conversion efficiency decreases in the high-temperature region and when there is a low angle of sunlight incidence on the solar cells, except for DSSCs [1,2].The exception occurs due to redox mediators (electrolytes).The solid nature of a redox mediator makes DSSCs scalable in manufacturing and eliminates the odds of a liquid or gel counterpart.The electrolyte participates in the following redox reactions: oxidation at the working electrode for dye regeneration, and reduction of the oxidized ions at the counter electrode.The reaction speed at the electrolyte/electrode interface largely controls DSSC efficiency.The speed depends on the electrical conductivity (σ) of the redox mediator.Therefore, a solid redox mediator (SRM) with σ 25 • C greater than 10 −4 S cm −1 is highly desirable for the fast movement of the redox couple.For a review, see references [4,5].
LiCF 3 SO 3 is one of the lithium salts that is highly used for preparing solid polymer electrolytes for lithium-ion batteries [45].The anion CF 3 SO 3 − is generally written as Triflate − for trifluoromethanesulfonic acid.Similar to the TFSI − ion, this anion coordinates weakly with a cation and therefore does not allow ion pairing [46].This anion has an ionic size of 0.44 nm, which is smaller to that of the TFSI − ion.LiTriflate possesses higher values of donor number and ionic mobility, and lower values of molecular weight and dissociation constant, than LiTFSI.LiTriflate, therefore, possesses a lower value of σ 25 • C in a mixture of solvents [42].In this study, we replaced the LiTFSI of [(1 − x)SN: xPEO]-LiTFSI-Co salts with LiTriflate to show the anionic effect on the electrical transport properties of SRMs.As previously mentioned, x = 0, 0.5, and 1 in the weight fraction.Hereafter, lithium salt is represented by LiX, where X = TFSI − or Triflate − .The molar composition and preparatory methods for the SRM [(1 − x)SN: xPEO]-LiX-Co salts are identical.We also prepared ACN-based liquid counterparts (LRMs) identically, as reported by Mathew et al. [44], for comparison.The preparation of the SRMs with x = 0 was identical to that of the LRMs.We used the conventional method (solution-cast) of preparation for the PEO-based SRMs (x = 0.5 and 1).The anionic effect on the electrical transport properties was explained using Fourier transform infrared spectroscopy (FT-IR), X-ray diffractometry (XRD), Xray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), UV-visible spectroscopy (UV-vis), differential scanning calorimetry (DSC), and thermogravimetric analysis (TGA).

Materials and Methods
The highly pure chemicals (cf.Table S2) were purchased and used without purification.The ACN-based LRMs and SN-based SRMs were synthesized using 0.1 M LiX, 0.25 M Co(bpy) 3 (TFSI) 2 , and 0.06 M Co(bpy) 3 (TFSI) 3 in ACN and SN, respectively, under stirring Polymers 2024, 16, 1436 3 of 18 at 65 • C for 24 h [3,44].The PEO-based SRMs with x = 0.5 and 1 were synthesized using the solution-cast method [32][33][34][35].The SRMs with x = 1 underwent a complete replacement of SN with PEO, followed by rigorous stirring in acetonitrile at 65 • C for 48 h, casting on a Teflon Petri dish, and drying under a nitrogen gas atmosphere at room temperature.The SRMs with x = 0.5 were synthesized similarly.The redox mediators were characterized using conventional techniques, such as impedance spectroscopy (IS), FT-IR, XRD, XPS, SEM, UV-vis, DSC, and TGA, which are described in the Supplementary Information (cf.Table S3) [3,47].Table S4 lists the equipment used for the measurements.

Results and Discussion
Impedance spectroscopy is a tool used to study the electrical transport properties of an electrolyte [48,49].A complex impedance plot, widely known as the Nyquist plot, helps to deduce bulk resistance (R b ) and, thereby, the electrical conductivity of the electrolyte.Figure 1 shows Nyquist curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts along with their liquid counterparts (LRMs) at 25 • C. Here, x = 0, 0.5, and 1, and X = TFSI − and Triflate − .In this figure, region (I) indicates a linear trend in the low-frequency domain because of the blocking-electrode effect, and region (II) corresponds to a semi-circle in the high-frequency domain because of the ionic diffusion effect [48,49].The Nyquist curve can be fitted using an equivalent circuit [49], R s,I + (R b ∥C 1 ) I + C 2,II , where the notations have their usual meaning.For example, R s stands for series resistance due to leads, R b for bulk resistance, C 1 for chemical capacitance, and C 2 for double-layer capacitance.The LRMs exhibited nearly identical and perfect trends: a linear curve in region (I) and a semi-circle in region (II).Having a plastic crystal phase, the SRMs with x = 0 had a semi-circle similar to those of SN-LiI-I 2 [21], but with a less prominent blocking-electrode effect.In addition, the semi-circle was slightly suppressed relative to those of the LRM-based redox mediators.The SRMs with x = 1 showed largely suppressed semi-circles, most probably because of the semi-crystalline nature of the PEO [50].The SRMs with x = 0.5 portrayed region (I) only.We found a similar pattern for the (SN-PEO)-MI-I 2 (M = Li + or K + ) SRMs [34,35].This indicated the formation of amorphous domains by the short and entangled polymer chains induced by the plasticizing properties of succinonitrile [51,52].Figure 1 also portrays the anionic effect.Relative to Triflate − , TFSI − resulted in a low value of bulk resistance, as marked by an arrow.This was associated with a smaller semi-circle and a prominent linear trend, except for the redox mediator with x = 0.
We evaluated the σ 25 • C values of liquid and solid redox mediators using the values of the R b , thickness, and area of the electrolyte.Table 1 lists the average values of σ 25 • C .The LRMs achieved σ 25 • C ≈ 1.7 × 10 −2 S cm −1 for X = TFSI − and ≈ 1.6 × 10 −2 S cm −1 for X = Triflate − , as reported earlier for liquid electrolytes [42].We recently attained σ 25 • C ≈10 −3 S cm −1 for SN-LiI-I 2 , because of the solid solvent nature of SN [21].We achieved similar σ 25 • C values, ≈2.1 × 10 −3 S cm −1 for X = TFSI − and ≈1.5 × 10 −3 S cm −1 for X = Triflate − for the SRMs with x = 0. Compared with the LRMs, these values are less than an order of magnitude.The complete replacement of SN by PEO decreased the σ 25 • C value to ≈9.7 × 10 −7 S cm −1 for X = TFSI − and ≈6.3 × 10 −7 S cm −1 for X = Triflate − , which are considerably less than three orders of magnitude.A poor σ 25 • C value was observed previously for several PEO-based solid I − /I − 3 redox mediators and occurred due to high PEO crystallinity, hindering ion transport [34,35].The redox mediators with x = 0.5 achieved a σ 25 • C value of ≈7.2 × 10 −4 S cm −1 for X = TFSI − and ≈3.1 × 10 −4 S cm −1 for X = Triflate − .These values are more than two orders of magnitude higher than those of the redox mediators with x = 1 and nearly an order of magnitude lower than those of the redox mediators with x = 0. We found similar values (3-7 × 10 −4 S cm −1 ) for the (PEO-SN) Blend-MI-I 2 (M = Li + and K + ) SRMs, too [34,35].This is due to the plasticizing nature of SN, which offers more amorphous regions for ion transport.One can note that TFSI − resulted in a better σ 25 • C value than Triflate − , specifically for the PEO-based redox mediators.This is due to several factors, such as the large ionic size; lower values of lattice energy, donor numbers, and ionic mobility; and higher values of the molecular weight and dissociation constant of TFSI − [42].Figure 2 shows the log σ − T −1 curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts (x = 0, 0.5, and 1; X = TFSI − or Triflate − ) and their liquid counterparts (LRMs).The SRMs with x = 0 and 1 showed a linear trend similar to the LRMs, SN-LiI-I2, and PEO-KI-I2, revealing Arrhenius-type behavior of ion transport in the form of σ = σo exp[-Ea/kBT] with the help of molecules or polymeric chains, where σo, Ea, and kB are the pre-exponential factor, activation energy, and Boltzmann constant, respectively [21,32,35].The SRMs with x = 0.5, on the other hand, had downward curves, just like the Blend-MI-I2.This showed Vogel-Tamman-Fulcher (VTF)-type behavior because of the formation of the amorphous phase.The VTF-type trend is expressed as σ = σoT −½ exp[-B/kB(T − To)], where B and To are the   Figure 2 shows the log σ − T −1 curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts (x = 0, 0.5, and 1; X = TFSI − or Triflate − ) and their liquid counterparts (LRMs).The SRMs with x = 0 and 1 showed a linear trend similar to the LRMs, SN-LiI-I 2 , and PEO-KI-I 2 , revealing Arrhenius-type behavior of ion transport in the form of σ = σ o exp[-E a /k B T] with the help of molecules or polymeric chains, where σ o , E a , and k B are the pre-exponential factor, activation energy, and Boltzmann constant, respectively [21,32,35].The SRMs with x = 0.5, on the other hand, had downward curves, just like the Blend-MI-I 2 .This showed Vogel-Tamman-Fulcher (VTF)-type behavior because of the formation of the amorphous phase.The VTF-type trend is expressed as σ , where B and T o are the pseudo-activation energy and free-volume temperature, respectively.The inset of Figure 2 exhibits a linear log(σT ½ ) -(T − T o ) −1 curve for SRMs with x = 0.5.We summarized the nature of redox mediators in Table 1.We also evaluated the values of E a and B from the slopes of the linear curves of redox mediators and listed them in Table 1 for comparison.In this table, regions I and II correspond to the temperatures before and after the melting temperature of SRMs with x = 0 and 1.Both the SRMs, x = 0 and 1, possessed E a -values of more than 0.3 eV in the solid-state region (region-I), making them useless for device applications [53].In contrast, the SRMs with x = 0.5 possessed quite a low B-value ≈ 0.06 eV, which was similar to those of Blend-MI-I 2 [34,35].Table 1 also shows lower activation energy values for the TFSI − -based redox mediators.As mentioned earlier, this is due to the several beneficial properties of TFSI − .
The level of interaction between the matrix and salt in the redox mediators is observed in the following order: LRM < x = 0 < 0.5 < 1.This order is the same for the anionic effect, too.We quantified the matrix-salt interaction by evaluating the relative intensity, ∆I = I 1105 /I 1196 , for both anions of redox mediators, where I 1105 stands for intensity at 1105 cm −1 for the ν s,COC mode of PEO, and I 1196 for the strongest peak at 1196 cm −1 for the ν a,CF3 mode of ionic salts.Figure 4 shows the relative intensity of LRMs and SRMs for both anions.The LRMs and SRMs (x = 0) possessed ∆I = 0 for both anions, indicating the least interaction between the solvent or matrix and ionic salt.The value of ∆I(X = TFSI − ) was ≈ 1.2 for the SRMs with x = 0.5 and 1, while the PEO-based SRMs with x = 0.5 and 1 had a higher value of ∆I(X = Triflate − ): 1.34 and 3.4, respectively.This revealed an increase in the level of interaction from x = 0.5 to 1, resulting in a hindrance in ion transport and a decrease in σ 25 • C value.This also demonstrated that ∆I(X = Triflate − ) is greater than ∆I(X = TFSI − ), indicating the anionic effect.The XRD study, discussed below, also supports this result.These clearly show the anionic effect, where TFSI − ions are more effective for C−H bond contraction and amorphous region formation, and therefore higher electrical conductivity.It is also noticeable that the ion pairing peaks neither were prominent nor appeared as shoulders; in fact, they overlapped with those of the PEO.
The level of interaction between the matrix and salt in the redox mediators is observed in the following order: LRM < x = 0 < 0.5 < 1.This order is the same for the anionic effect, too.We quantified the matrix-salt interaction by evaluating the relative intensity, ∆ =   ⁄ , for both anions of redox mediators, where  stands for intensity at 1105 cm −1 for the νs,COC mode of PEO, and  for the strongest peak at 1196 cm −1 for the νa,CF3 mode of ionic salts.Figure 4 shows the relative intensity of LRMs and SRMs for both anions.The LRMs and SRMs (x = 0) possessed Δ = 0 for both anions, indicating the least interaction between the solvent or matrix and ionic salt.The value of Δ(X = TFSI − ) was ≈ 1.2 for the SRMs with x = 0.5 and 1, while the PEO-based SRMs with x = 0.5 and 1 had a higher value of Δ(X = Triflate − ): 1.34 and 3.4, respectively.This revealed an increase in the level of interaction from x = 0.5 to 1, resulting in a hindrance in ion transport and a decrease in σ25°C value.This also demonstrated that Δ(X = Triflate − ) is greater than Δ(X = TFSI − ), indicating the anionic effect.The XRD study, discussed below, also supports this result.).The patterns of these mediators had no peaks for other ingredients.In contrast, the redox mediators with x = 0.5 exhibited no peaks.An amorphous peak appeared at 27.6 • for X = TFSI − and 23.1 • for X = Triflate − .These indicate disorder in the SN molecules, eutectic phase formation, and amorphous phase formation in the PEO or blend matrix [3,21,[33][34][35]63].For the blend-based redox mediators (x = 0.5), PEO acted as an impurity, abolishing the crystalline structure of the SN in the presence of ions [3,[33][34][35].The absence of reflection peaks of the ionic salts also confirmed complete dissolution and complexation of the salts.The anionic effect is visible for all the redox mediators, x = 0 − 1.The redox mediator with x = 0 exhibited relatively stronger reflection peaks of succinonitrile for X = TFSI − than Triflate − .The latter is indicative of eutectic phase formation, as indicated by the DSC study, which is discussed later [21,63].The redox mediator (x = 1) had stronger reflection peaks of PEO for X = Triflate − than TFSI − , revealing a higher level of PEO crystallinity and, thereby, a lower value of electrical conductivity.The SEM results, discussed later, corroborate these results.For the blend-based redox mediators (x = 0.5), PEO acted as an impurity, abolishing the crystalline structure of the SN in the presence of ions [3,[33][34][35].The absence of reflection peaks of the ionic salts also confirmed complete dissolution and complexation of the salts.The anionic effect is visible for all the redox mediators, x = 0 − 1.The redox mediator with x = 0 exhibited relatively stronger reflection peaks of succinonitrile for X = TFSI − than Triflate − .The latter is indicative of eutectic phase formation, as indicated by the DSC study, which is discussed later [21,63].The redox mediator (x = 1) had stronger reflection peaks of PEO for X = Triflate − than TFSI − , revealing a higher level of PEO crystallinity and, thereby, a lower value of electrical conductivity.The SEM results, discussed later, corroborate these results.X-ray photoelectron spectroscopy is a tool used to study the interaction between ingredients at the surface [47,[64][65][66][67][68]. Figure S2 shows the survey spectra of SRM [(1 − x)SN: xPEO]-LiX-Co salts (x = 0, 0.5, and 1; X = TFSI − and Triflate − ).The survey spectrum was corrected to fix the C 1 s peak at 284.6 eV [64][65][66].The survey spectra depicted S 2p, C 1 s, N 1 s, O 1 s, and F 1 s elements for all redox mediators.However, only the redox mediators with x = 0 exhibited Ti 2p peaks.The presence of this peak can be attributed to the enhanced penetration of SN-based electrolytes into the pores of the TiO2 layer and the creation of a remarkably thin film on TiO2, facilitated by the solid solvent characteristic of SN.The SEM image demonstrates the infiltration and development of a thin coating of SN-   X-ray photoelectron spectroscopy is a tool used to study the interaction between ingredients at the surface [47,[64][65][66][67][68]. Figure S2 shows the survey spectra of SRM [(1 − x)SN: xPEO]-LiX-Co salts (x = 0, 0.5, and 1; X = TFSI − and Triflate − ).The survey spectrum was corrected to fix the C 1 s peak at 284.6 eV [64][65][66].The survey spectra depicted S 2p, C 1 s, N 1 s, O 1 s, and F 1 s elements for all redox mediators.However, only the redox mediators with x = 0 exhibited Ti 2p peaks.The presence of this peak can be attributed to the enhanced penetration of SN-based electrolytes into the pores of the TiO 2 layer and the creation of a remarkably thin film on TiO 2 , facilitated by the solid solvent characteristic of SN.The SEM image demonstrates the infiltration and development of a thin coating of SN-based electrolyte on the TiO 2 substrate, as described in the next paragraph.The TiO 2 substrate is anticipated to exhibit a substantial accumulation of PEO-based redox mediators (x = 0.5 and 1), hence impeding the examination of the electrolyte/TiO 2 interface.The peak's position, intensity, and width (full width at half maximum) were determined through the fitting of the smoothed and baseline-corrected spectrum.Figures S3-S5 depict the best-fit plots of different elements of SRMs with x = 0, 0.5, and 1, respectively.The left and right columns are for SRMs, with X = TFSI − and Triflate − , respectively.Figure 6 shows smoothed and baseline-corrected XPS spectra of the S 2p, C 1 s, N 1 s, O 1 s, and F 1 s elements for the SRM [(1 − x)SN: xPEO]-LiX-Co salts with x = (a) 0, (b) 0.5, and (c) 1, where X = TFSI − and Triflate − .Figure 6a also portrays the spectra of the Ti 2p element.The peaks can be assigned in light of previously reported studies [47,[64][65][66][67][68].The SRMs depicted a small-height S 2p peak because of the −SO 2 − group at ≈168.2 eV for the spin of 3/2, associated with a shoulder peak at ≈169.8 eV for the spin of ½.The area of the C 1 s core level is the most complex.This area showed a standard mid-height peak at 284.6 eV for the alkyl (−C−C−) group associated with a strong peak at ≈286 eV for the bpy ring or −C−C−O− group, a small-height peak at ≈288.6 eV for the −C≡N group, and a distinctive small-height peak at ≈292 eV for the -CF 3 group.The N 1 s spectrum depicted a small-height broad peak consisting of two deconvoluted peaks because of anions at ≈398 eV and the bpy ring at ≈400 eV.The Ti 2p spectrum of the TiO 2 layer exhibited two strong and distinctive peaks at ≈458.2 eV and ≈464 eV because of the spin-orbit-coupling phenomenon.The SRMs with x = 0 portrayed two O 1 s core-level strong distinctive peaks at ≈529.8 eV and ≈532 eV because of the −SO 2 − group in the anions.The PEO-based SRMs (x = 0.5 and 1) exhibited the strongest O 1 s peak with deconvolution of ≈531 eV and ≈532 eV because of the spin-orbit-coupling phenomenon.The redox mediators also showed a strong F 1 s core-level distinctive peak at ≈688.6 eV because of the -CF 3 group of anions.We divided the intensity by the width to make the ratio dimensionless.Figure 7 plots the intensity/width (=R) against the peak position for all redox mediators with x = 0, 0.5, and 1.The solid symbols represent redox mediators with X = TFSI − , and open symbols denote redox mediators with X = Triflate − .This figure demonstrates a shift either in the ratio (∆R = R TFSI − R Triflate ; cf. Figure S6) or peak position (∆P = Peak Position TFSI − Peak Position Triflate ; cf. Figure S7) or both, revealing an anionic effect.The level of shift differed depending on the composition and element.For example, the SRMs with x = 0 showed ∆R < 0, except for O 1 s, where ∆R > 0, too; the redox mediators with x = 0.5 depicted ∆R > 0, except for O 1 s and F 1 s, where ∆R < 0; and the redox mediators with x = 1 exhibited both positive and negative ∆R.At F 1 s, the negative shift was minimum for x = 0.5 and maximum for x = 0.The SRMs with x = 0 showed ∆P > 0 for S 2p, C 1 s, O 1 s, and F 1 s core levels, and ∆P < 0 for N 1 s.The redox mediators with x = 0.5 exhibited ∆P ≥ 0 for S 2p, C 1 s, and F 1 s core levels, and ∆P < 0 for N 1 s and O 1 s.The redox mediators with x = 1 depicted ∆P ≥ 0 for all core levels.
Polymers 2024, 16, 1436 12 of 18 however, blending SN and PEO largely improved the transmittance.This is due to a decrease in PEO crystallinity because of the plasticizing properties of the SN [35].Table 2 portrays the anionic effect, too, where TFSI − led to better transmittance than Triflate − .This is due to a decrease in PEO crystallinity, as pointed out by the DSC results, which are discussed below.Figure 10 shows the DSC curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts, where x = 0, 0.5, and 1, and X = TFSI − and Triflate − .The SRMs with x = 0 portrayed two endothermic peaks marked by Tpc and Tm for crystal-to-plastic-crystal phase transition temperature and melting temperature, respectively.Table 3 lists the values of Tpc and Tm for comparison.For the SN matrix, Tpc = -38.4°C and Tm = 57.7 °C [21,63].It is worth mentioning that the area of Tm-peak corresponds to the heat enthalpy or crystallinity of the redox mediator [21,32].We observed that the position and area of Tm-peak decreased for the SRMs with x = 0 relative to the pure SN matrix.This is indicative of a decrease in the crystallinity of succinonitrile [21].The Tpc-peak showed a position similar to that of the pure matrix, as observed earlier for the SN-LiI-I2 redox mediator, however, with an increase in the area, most probably because of the SN-ionic salt interaction [21,63].The anionic effect is also noticeable for SRMs with x = 0.For example, TFSI − resulted in a broad Tm-peak, which is indicative of the disordered plastic crystalline nature of SN.In contrast, Triflate − yielded multiplets at higher temperatures with a larger area, indicating eutectic phase formation along with the disordered plastic crystal phase of SN.TFSI − also resulted in the area (40.9) of the Tpc-peak being less than that for Triflate − (139.9), which is indicative of less crystallinity and, thereby, higher electrical conductivity in the TFSI − -based redox mediator.The SRMs with x = 1 had a Tm-peak with values of only 63.8 °C for X = TFSI − and 65.2 °C for X = Triflate − , which are less than the Tm-value of the PEO matrix (65.7 °C [32]).The area of the Tm-peak was also less than that of the PEO matrix, revealing a decrease in PEO crystallinity.These DSC curves also exhibited the anionic effect via a decrease in the position and area of the Tm-peak for X = TFSI − , revealing a decrease in PEO crystallinity and, thereby, an increase in electrical conductivity [32][33][34][35].The SRMs with x = 0.5 possessed negligibly small and broad Tm-peaks with values of only 4 °C for X = TFSI − and 7.8-26 °C   [21,63].It is worth mentioning that the area of T m -peak corresponds to the heat enthalpy or crystallinity of the redox mediator [21,32].We observed that the position and area of T m -peak decreased for the SRMs with x = 0 relative to the pure SN matrix.This is indicative of a decrease in the crystallinity of succinonitrile [21].The T pc -peak showed a position similar to that of the pure matrix, as observed earlier for the SN-LiI-I 2 redox mediator, however, with an increase in the area, most probably because of the SN-ionic salt interaction [21,63].The anionic effect is also noticeable for SRMs with x = 0.For example, TFSI − resulted in a broad T m -peak, which is indicative of the disordered plastic crystalline nature of SN.In contrast, Triflate − yielded multiplets at higher temperatures with a larger area, indicating eutectic phase formation along with the disordered plastic crystal phase of SN.TFSI − also resulted in the area (40.9) of the T pc -peak being less than that for Triflate − (139.9), which is indicative of less crystallinity and, thereby, higher electrical conductivity in the TFSI − -based redox mediator.The SRMs with x = 1 had a T m -peak with values of only 63.8 • C for X = TFSI − and 65.2 • C for X = Triflate − , which are less than the T m -value of the PEO matrix (65.7 • C [32]).The area of the T m -peak was also less than that of the PEO matrix, revealing a decrease in PEO crystallinity.These DSC curves also exhibited the anionic effect via a decrease in the position and area of the T m -peak for X = TFSI − , revealing a decrease in PEO crystallinity and, thereby, an increase in electrical conductivity [32][33][34][35].
Polymers 2024, 16, 1436 13 of 18 The SRMs with x = 0.5 possessed negligibly small and broad T m -peaks with values of only 4 • C for X = TFSI − and 7.8-26 • C for X = Triflate − .These values, as well as the peak area, are less than those of the PEO-SN blend matrix (T m ≈ 30.1 • C [32]), indicating a decrease in PEO crystallinity.The anionic effect was visualized at the T m -peak position and area.TFSI − led to lower values of T m and a smaller area compared to Triflate − , resulting in lower PEO crystallinity and, thereby, a higher σ 25 • C value.These results are also consistent with the findings of the TGA investigation, which are discussed below.
blend matrix (Tm ≈ 30.1 °C [32]), indicating a decrease in PEO crystallinity.The anionic effect was visualized at the Tm-peak position and area.TFSI − led to lower values of Tm and a smaller area compared to Triflate − , resulting in lower PEO crystallinity and, thereby, a higher σ25°C value.These results are also consistent with the findings of the TGA investigation, which are discussed below.Figure 11 shows the TGA curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts with x = 0, 0.5, and 1 for X = TFSI − and Triflate − .The initial plateau region of the curve corresponds to the thermal stability of the SRM, which is ≈75 °C for x = 0, 125 °C for x = 0.5, and 200 °C for x = 1.These values are similar to those of the pure matrices [32].The SRMs with x = 0 and 1 exhibited one-step dropping due to the decomposition of the matrix, while the mediators with x = 0.5 showed two-step dropping due to the decomposition of the SN and PEO matrices [3,32].These curves also portrayed an anionic effect via the degradation phenomenon.TFSI − resulted in more rapid degradation compared to Triflate − , particularly for SRMs with x = 0.5 and 1.This is because TFSI − -based redox mediators have lower PEO crystallinity.  Figure 11 shows the TGA curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts with x = 0, 0.5, and 1 for X = TFSI − and Triflate − .The initial plateau region of the curve corresponds to the thermal stability of the SRM, which is ≈75 • C for x = 0, 125 • C for x = 0.5, and 200 • C for x = 1.These values are similar to those of the pure matrices [32].The SRMs with x = 0 and 1 exhibited one-step dropping due to the decomposition of the matrix, while the mediators with x = 0.5 showed two-step dropping due to the decomposition of the SN and PEO matrices [3,32].These curves also portrayed an anionic effect via the degradation phenomenon.TFSI − resulted in more rapid degradation compared to Triflate − , particularly for SRMs with x = 0.5 and 1.This is because TFSI − -based redox mediators have lower PEO crystallinity.As mentioned earlier, an SRM with σ25°C > 10 −4 S cm −1 and Ea < 0.3 eV is required to regenerate dye molecules faster at the TiO2/dye/electrolyte interface and, thereby, inhibit back-electron transfer [6][7][8][9][10][11][12][13][14][15][16][17][18][19].Faster dye regeneration via the faster oxidation of ionic species results in a higher photocurrent and cell efficiency.This also results in faster ionic species regeneration.However, the higher mass of metal ions, e.g., Co 2+ /Co 3+ , slows down their diffusion in DSSCs, which decreases the concentration of metal ions at the TiO2/dye/electrolyte interface, resulting in poor cell efficiency [69,70].Hence, it is imperative to enhance the porosity of the mesoporous TiO2 film in order to accommodate a greater number of ionic species at the interface [70].It is also necessary to have TiO2 pores that are water-free because water changes the surface of TiO2 and decreases cell efficiency [4,71].In the present scenario, due to their superior electrical transport and optical properties, the DSSCs with TFSI − -based SRMs are expected to perform better than those with Triflate − -based SRMs.However, porosity at the TiO2/dye/electrolyte interface needs to be optimized by utilizing a mixture of large and small anatase TiO2 nanoparticles [69,70,[72][73][74].Therefore, we will perform DSSC fabrication and characterizations in the future.

Conclusions
We studied the electrical transport properties of SRM [(1 − x)SN: xPEO]-LiX-Co salts (x = 0, 0.5, and 1; X = TFSI − and Triflate − ), and the results were compared with those of their acetonitrile-based liquid counterparts (LRMs).The LRMs exhibited σ25°C ≈ 10 −2 S cm −1 .The SRMs achieved σ25°C ≈ 10 −3 S cm −1 for x = 0, >10 −4 S cm −1 for x = 0.5, and ≈10 −6 S cm −1 for x = 1.In these redox mediators, σ25°C(TFSI − ) > σ25°C(Triflate − ) and Ea(TFSI − ) < Ea(Triflate − ).FT-IR spectroscopy showed the interaction and, thereby, the anionic effect in the following order: LRM < 0 < 0.5 < 1.The XRD study exhibited an increase in PEO crystallinity from x = 0.5 to 1, which was more for Triflate − ions.The XPS study showed a shift in the intensity/width ratio and peak position of the elements because of the anionic effect.The SEM images depicted an increase in surface roughness from x = 0.5 to 1, which was higher for Triflate − ions.The transmittance showed the following orders: LRM = 0 (= 0%) << 1 << 0.5 in the UV-A region, 0 << 1 << LRM ≈ 0.5 in the visible region, and 0 << 1 < 0.5 < LRM in the As mentioned earlier, an SRM with σ 25 • C > 10 −4 S cm −1 and E a < 0.3 eV is required to regenerate dye molecules faster at the TiO 2 /dye/electrolyte interface and, thereby, inhibit back-electron transfer [6][7][8][9][10][11][12][13][14][15][16][17][18][19].Faster dye regeneration via the faster oxidation of ionic species results in a higher photocurrent and cell efficiency.This also results in faster ionic species regeneration.However, the higher mass of metal ions, e.g., Co 2+ /Co 3+ , slows down their diffusion in DSSCs, which decreases the concentration of metal ions at the TiO 2 /dye/electrolyte interface, resulting in poor cell efficiency [69,70].Hence, it is imperative to enhance the porosity of the mesoporous TiO 2 film in order to accommodate a greater number of ionic species at the interface [70].It is also necessary to have TiO 2 pores that are water-free because water changes the surface of TiO 2 and decreases cell efficiency [4,71].In the present scenario, due to their superior electrical transport and optical properties, the DSSCs with TFSI − -based SRMs are expected to perform better than those with Triflate − -based SRMs.However, porosity at the TiO 2 /dye/electrolyte interface needs to be optimized by utilizing a mixture of large and small anatase TiO 2 nanoparticles [69,70,[72][73][74].Therefore, we will perform DSSC fabrication and characterizations in the future.

Figure 5
Figure 5 shows the XRD patterns of the SRM [(1 − x)SN: xPEO]-LiX-Co salts with x = 0, 0.5, and 1 for X = TFSI − (solid line) and Triflate − (dotted line).Table2lists the observed peaks of redox mediators with x = 0 and 1, which are small and broad as well

Figure 10
Figure10shows the DSC curves of the SRM [(1 − x)SN: xPEO]-LiX-Co salts, where x = 0, 0.5, and 1, and X = TFSI − and Triflate − .The SRMs with x = 0 portrayed two endothermic peaks marked by T pc and T m for crystal-to-plastic-crystal phase transition temperature and melting temperature, respectively.Table3lists the values of T pc and T m for comparison.For the SN matrix, T pc = -38.4• C and T m = 57.7 • C[21,63].It is worth mentioning that the area of T m -peak corresponds to the heat enthalpy or crystallinity of the redox mediator[21,32].We observed that the position and area of T m -peak decreased for the SRMs with x = 0 relative to the pure SN matrix.This is indicative of a decrease in the crystallinity of succinonitrile[21].The T pc -peak showed a position similar to that of the pure matrix, as observed earlier for the SN-LiI-I 2 redox mediator, however, with an increase in the area, most probably because of the SN-ionic salt interaction[21,63].The anionic effect is also noticeable for SRMs with x = 0.For example, TFSI − resulted in a broad T m -peak, which is indicative of the disordered plastic crystalline nature of SN.In contrast, Triflate − yielded multiplets at higher temperatures with a larger area, indicating eutectic phase formation along with the disordered plastic crystal phase of SN.TFSI − also resulted in the area (40.9) of the T pc -peak being less than that for Triflate − (139.9), which is indicative of less crystallinity and, thereby, higher electrical conductivity in the TFSI − -based redox mediator.The SRMs with x = 1 had a T m -peak with values of only 63.8 • C for X = TFSI − and 65.2 • C for X = Triflate − , which are less than the T m -value of the PEO matrix (65.7 • C[32]).The area of the T m -peak was also less than that of the PEO matrix, revealing a decrease in PEO crystallinity.These DSC curves also exhibited the anionic effect via a decrease in the position and area of the T m -peak for X = TFSI − , revealing a decrease in PEO crystallinity and, thereby, an increase in electrical conductivity[32][33][34][35].

Table 2
slightly shifted relative to the matrix (SN: 2θ ≈ 19.7 • and 28.1 • ; PEO: 2θ ≈ 19.2 • and 23.3 • [32] lists the observed peaks of redox mediators with x = 0 and 1, which are small and broad as well as